Você está na página 1de 8

Original Research

Rapid Treatment Response of Suicidal Symptoms to Lithium,


Sleep Deprivation, and Light Therapy (Chronotherapeutics)
in Drug-Resistant Bipolar Depression
Francesco Benedetti, MD; Roberta Riccaboni, PsyD; Clara Locatelli, MD;
Sara Poletti, PhD; Sara Dallaspezia, MD; and Cristina Colombo, MD

ABSTRACT
Background: One third of patients with bipolar disorder attempt
B ipolar depression is a difficult-to-treat condition that
has been associated with extremely low success rates of
antidepressant drugs in naturalistic settings.1 Patients with
suicide. Depression in bipolar disorder is associated with drug bipolar disorder spend a substantial proportion of their time
resistance. The efficacy of antidepressants on suicidality has been ill,2 with depression representing their predominant abnormal
questioned. Total sleep deprivation and light therapy prompt a mood state,3 and with the repeated use of antidepressant drugs
rapid and stable antidepressant response in bipolar disorder. being related to poor prospective response to naturalistic
Method: We studied 143 consecutively admitted inpatients antidepressant treatment.4 Possibly as a consequence of
(December 2006–August 2012) with a major depressive episode their disabling condition, about 30% of patients with bipolar
in the course of bipolar disorder (DSM-IV criteria). Among the disorder attempt suicide,5,6 and about 20% eventually die of
141 study completers, 23% had a positive history of attempted suicide.7,8
suicide and 83% had a positive history of drug resistance. During Treatment of suicidality is a major issue, but few options are
1 week, patients were administered 3 consecutive total sleep available. Patients with suicidal thoughts or intent compared to
deprivation cycles (each composed of a period of 36 hours those without suicidal thoughts or intent have a clearer lifetime
awake followed by recovery sleep) combined with bright light history of recurrence of major depressive episodes but are
therapy in the morning for 2 weeks. At admission, patients usually excluded from trials,9 thus limiting the generalizability
who had been taking lithium continued it, and those who
of results.10 Antidepressant interventions prompt remission
had not been taking lithium started it. Severity of depression
was rated according to the Hamilton Depression Rating Scale
and can then reduce the suicide risk associated with acute
(HDRS) (primary outcome measure) and Beck Depression mood episodes, but a large-scale epidemiologic study11 did
Inventory (BDI). not support the usefulness of antidepressant drugs in reducing
lifetime completed and attempted suicide in mood disorders.
Results: Two patients switched polarity. Among the 141 who
completed the treatment, 70% achieved a 50% reduction in
Administration of antidepressants to acutely suicidal patients
HDRS score in 1 week, which persisted 1 month after in 55%. can even be risky because of an age-dependent risk of suicidal
The amelioration involved an immediate and persistent decrease behavior and ideation associated with use of antidepressants,
in suicide scores soon after the first total sleep deprivation cycle which led to a US Food and Drug Administration black box
(F3,411 = 42.78, P < .00001). A positive history of suicide attempts warning for patients under age 25 years.12 When effective,
was associated with worse early life stress and with worse suicide antidepressants are slow, however, with literature trials
scores at baseline, but it did not influence response. Patients with showing that no difference can be expected during the first 2
current suicidal thinking or planning responded equally well weeks between active and placebo treatments13 Additionally,
(F3,42 = 20.70, P < .000001). Remarkably, however, nonresponders the selective publication of positive trials boosted the apparent
achieved a benefit, with significantly decreased final scores also efficacy of active drugs.14 Lithium is able to reduce the lifetime
including suicidality ratings (F3,120 = 6.55, P = .0004). Self-ratings
suicide risk of patients with bipolar disorder to the same levels
showed the same pattern of change. Previous history of drug
resistance did not hamper response. During the following
of the general population,15,16 but it shows a long latency of
month, 78 of 99 responders continued to stay well and were antidepressant action and is then of little help in the acute
discharged from the hospital on lithium therapy alone. phase. Finally, the overall depression severity is moderately
associated with suicidal ideation, which associates more with
Conclusions: The combination of total sleep deprivation,
light therapy, and lithium is able to rapidly decrease depressive
core mood symptoms and self-punitive thinking, and can
suicidality and prompt antidepressant response in drug-resistant remain high even when somatic and vegetative symptoms
major depression in the course of bipolar disorder. improve as a result of treatment response.17,18
J Clin Psychiatry
The unmet clinical need for the rapid resolution of
© Copyright 2013 Physicians Postgraduate Press, Inc. breakthrough life-threatening symptoms,19 such as suicide,
and the conundrum between the risks of treatment-emergent
mania20,21 and of relapse after treatment discontinuation22 often
Submitted: March 6, 2013; accepted June 24, 2013.
Online ahead of print: December 10, 2013 (doi:10.4088/JCP.13m08455). lead to prolonged and highly complex medication regimens
Corresponding author: Francesco Benedetti, MD, Istituto Scientifico to achieve a stable response in bipolar disorder.23 Hence, the
Universitario Ospedale San Raffaele, Dipartimento di Neuroscienze
Cliniche, San Raffaele Turro, Via Stamira d’Ancona 20, 20127 Milan, Italy interest in chronotherapeutics that might eliminate the long
(benedetti.francesco@hsr.it). latencies of traditional antidepressant treatments yet offer

© 2013 COPYRIGHT PHYSICIANS POSTGRADUATE PRESS, INC. NOT FOR DISTRIBUTION, DISPLAY, OR COMMERCIAL PURPOSES.
J Clin Psychiatry e1
Benedetti et al

between 0800 and 0900. Light therapy in the morning was


■■ A rapid treatment of acute suicidality is an unmet need in then continued for 2 weeks. At admission, patients who
severe drug-resistant bipolar depression because the efficacy had been taking lithium continued it (n = 49), and those
Clinical Points

of antidepressants on suicidality has been questioned and who had not been taking lithium started it together with
because clearly effective drugs, such as lithium, have a long the chronotherapeutic procedure to enhance its effect and
latency of action. prevent relapse (n = 92).34–36 No other antidepressant was
■■ The combination of chronotherapeutic techniques (sleep administered.
deprivation, light therapy) with lithium can prompt Patients were followed up for 1 month after the acute
treatment response in drug-resistant depression, with a chronotherapeutic treatment. Nonresponders were treated
clinical benefit that involves an immediate decrease in by the psychiatrists in charge upon clinical need. Responders
suicidality. These benefits partly extend to nonresponders. continued lithium alone.

Data Collection and Analysis


comparable efficacy.24–27 Previous studies28–30 by our group Severity of depression was rated (days 0, 1, 2, and 6) by the
showed that the combination of lithium, repeated total sleep psychiatrists in charge of the patients according to a modified
deprivation, and light therapy caused a stable remission of version of the 21-item HDRS (referred to as the HDRS-
bipolar depression in more than half of the treated patients. NOW,37 the primary outcome measure), from which items
Despite suicidal thoughts or intent not having been an that could not be meaningfully rated due to the total sleep
exclusion criterion in these trials, the specific issue of the deprivation procedure were excluded (ie, weight changes
efficacy of chronotherapeutics with regard to suicidality was and insomnia: item numbers 4, 5, 6, and 16). Response
never addressed. Thus, in the present study, we aimed to was defined as a 50% reduction of HDRS scores. Perceived
investigate a homogenous sample of patients affected by a depression was self-rated on the 13-item Beck Depression
major depressive episode in the course of bipolar disorder. Inventory (BDI).38 Current suicidality was rated according
to the HDRS suicide item, which correlates with number of
METHOD suicide attempts, age at first attempt, and other ratings of
Patients and Treatment suicidality39,40 and on the BDI.
We studied 143 consecutively admitted inpatients Given the negative relationship between response to
affected by a major depressive episode without psychotic total sleep deprivation and history of drug resistance,29 the
features in the course of bipolar disorder (DSM-IV criteria, latter was assessed according to Thase and Rush criteria41:
SCID-I interview31). Among the 141 study completers, 32 24 patients (17.0%) had no history of drug resistance, 54
patients (22.7%) had a positive history of suicidality. The (38.3%) were resistant to 1 class of drugs, 42 (29.8%) were
criterion for suicidality was 1 or more documented suicide resistant to 2 classes, and 21 (14.9%) had higher stages of
attempt (broadly defined as any behavior aimed at killing resistance.
oneself) during the lifetime.32 This sample does not overlap Given the positive relationship between early life stress
with those described in previous studies.29,30 Patients were and suicidality in bipolar disorder,6 early (age 5–15 years) and
recruited between December 2006 and August 2012. recent (last 3 years) stressful life events were scored according
Inclusion criteria were a baseline Hamilton Depression to the Social Readjustment Rating Scale (SRRS),42 which
Rating Scale (HDRS)33 score of 18 or higher; absence of focuses on occurrences that lead to readjustment-requiring
other Axis I diagnoses; absence of mental retardation in changes in usual activities43 and which was validated in
Axis II diagnoses; absence of pregnancy, history of epilepsy, similar settings.32
and major medical and neurologic disorders; no treatment Repeated-measures analyses of variance were performed
with long-acting neuroleptic drugs in the last 3 months in the context of the general linear model.44,45 The analysis
before admission; and absence of a history of drug or was separately performed on the 3 outcome measures:
alcohol dependency or abuse within the last 6 months. After HDRS, HDRS item 3 (suicide) score, and BDI. The main
complete description of the study to the subjects, written factors of interest were response to treatment and history of
informed consent was obtained. The study was approved by suicide attempts. Time and ongoing lithium treatment were
the local ethics committee. also considered as factors. The significance of the effect of
All patients were administered 3 consecutive total sleep the single independent factors on the dependent variable was
deprivation cycles (day 0–7); each cycle was composed of a estimated (least squares method) by parametric estimates of
period of 36 hours awake. On days 0, 2, and 4, patients were predictor variables and following standard computational
totally sleep deprived from 0700 until 1900 of the following procedures.46 Analyses were separately performed in the
day. They were then allowed to sleep during the night in a whole sample and in patients who presented current suicide
sleep window 1900–0800 of days 1, 3, and 5. Patients were thinking or planning.9,10
administered light therapy (exposure for 30 minutes to a
10,000 lux bright white light, color temperature 4,600 K) RESULTS
at 0300 during the total sleep deprivation night and in the One hundred forty-one patients completed the treatment.
morning after recovery sleep, half an hour after awakening, Two patients switched polarity, showing moderate manic

© 2013 COPYRIGHT PHYSICIANS POSTGRADUATE PRESS, INC. NOT FOR DISTRIBUTION, DISPLAY, OR COMMERCIAL
e2
PURPOSES.
J Clin Psychiatry
Efficacy of Chronotherapy on Suicidality

Table 1. Clinical and Demographic Characteristics of the 141 Participants, Divided According to Their Lifetime
History of Suicide Attempts and Response to Antidepressant Treatmenta
Positive History Negative History
of Attempted Suicide of Attempted Suicide
Characteristic Responders Nonresponders Responders F or χ2
Nonresponders P
n 21 11 78 31 0.42 NS
Age, y 45.71 (11.90) 43.36 (8.61) 47.91 (10.97) 47.61 (13.44)
0.63 NS
Sex, n 2.41 NS
Male 6 5 29 15
Female 15 6 49 16
Education, y 11.38 (3.88) 12.73 (3.58) 10.8 (3.94) 11.87 (5.19) 0.91 NS
Age at onset, y 32.71 (10.77) 23.27 (5.73) 31.77 (10.59) 33.10 (9.16) 2.85 NS
Duration of illness, y 13.00 (8.36) 20.09 (9.59) 16.14 (9.22) 14.52 (12.95) 1.39 NS
Duration of current episode, wk 33.18 (40.49) 12.13 (7.43) 22.19 (21.31) 29.74 (31.20) 1.59 NS
No. of previous depressive episodes 5.00 (4.27) 3.82 (2.18) 5.99 (5.94) 6.55 (7.66) 0.71 NS
No. of previous manic episodes 3.10 (3.10) 1.91 (1.22) 3.88 (4.82) 3.10 (4.02) 0.85 NS
Total no. of previous recurrences 8.10 (6.56) 5.73 (3.23) 9.87 (10.03) 9.66 (10.76) 0.75 NS
Early stressors, no. of eventsb 15.54 (10.94) 20.30 (11.03) 12.78 (9.10) 11.09 (6.73) 2.70 < .05
SRRS scoreb 345.08 (249.46) 450.20 (272.06) 301.11 (195.42) 235.55 (125.64) 2.84 < .05
Recent stressors, no. of events 19.54 (8.13) 21.00 (9.30) 17.17 (11.53) 16.14 (10.03) 0.65 NS
SRRS score 506.85 (201.99) 537.90 (283.88) 462.13 (306.39) 393.09 (212.99) 0.83 NS
21-item HDRS baseline score 23.76 (3.95) 21.60 (3.20) 23.27 (4.80) 23.59 (4.12) 0.61 NS
HDRS-NOW day 1 20.71 (3.93) 18.45 (2.77) 20.01 (4.51) 20.45 (4.75) 0.73 NS
HDRS-NOW day 2c 10.24 (5.99) 16.18 (3.84) 11.68 (6.07) 16.29 (6.63) 6.71 < .001
HDRS-NOW day 3c 9.95 (5.47) 16.36 (3.56) 9.83 (5.90) 15.58 (5.29) 11.23 < .001
HDRS-NOW day 7c 4.33 (3.26) 15.00 (5.40) 4.37 (3.13) 16.19 (4.90) 90.59 < .001
BDI baseline scorec 15.67 (7.86) 21.64 (8.76) 14.49 (7.34) 19.00 (7.63) 4.40 < .01
aValues shown are mean (SD) unless stated otherwise.
bSignificantly higher values in patients with a positive history of suicide attempts.
cSignificantly lower values in responders.

Abbreviations: BDI = Beck Depression Inventory, HDRS-NOW = modified version of Hamilton Depression Rating Scale,
NS = nonsignificant, SRRS = Social Readjustment Rating Scale.

symptoms after the first total sleep deprivation, and Figure 1. Pattern of Change of Depression Severity (HDRS
were then treated with benzodiazepines and a second score) During Treatment in Patients Divided According to
mood stabilizer, thus restoring euthymia.47 Clinical and Chronotherapeutic Treatment Response and to Previous History
demographic characteristics of the 141 completers, divided of Suicide Attemptsa
according to history of suicide attempts and response to
p on

Re lee p on

p on

Re lee p on

Re lee p on

p on
ee ti

ee ti

ee ti

ee ti

ee ti

ee ti
ta y Sl riva

ta y Sl riva

Sl iva

ta y Sl riva

ta y Sl riva

Sl iva
treatment, are summarized in Table 1.
ry epr

ry epr
r ep

r ep

r ep
e
co ep D

co ep D

co p D

co ep D

co p D

co p D
Patients with a positive history of suicide attempts
e

e
ve

ve

ve

ve

ve

ve
Sl

Sl
lS

lS

lS

lS
al

al

compared to those with a negative history showed higher


t

t
Re

Re

Re
To

To

To

To

To

To
24
levels of early life stress (mean [SD] number of events: 22
17.61 [10.99] vs 12.29 [8.47], t = 2.91, P = .004; SRRS 20
score: 390.78 [258.97] vs 282.13 [179.73]; Student t = 2.70, 18
P = .008) but not of recent stress. Responders showed a 16
HDRS Score, Mean

14
lower severity of current depression than nonresponders
12
on baseline self-rated BDI score (14.71 [7.37] vs 19.59
10
[7.80]; t = 3.25, P = .001) but not on HDRS. No other effect 8
of these variables, nor their interaction, was significant. 6
4
Effect of Chronotherapeutics 2 Positive history (n = 21) Positive history (n = 11)
Negative history (n = 78) Negative history (n = 31)
Ninety-nine patients (70.1%) responded to treatment. 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
A positive history of suicide attempts was not associated
Time, d Time, d
with response (50% reduction in HDRS score; see Table Responders (n = 99) Nonresponders (n = 42)
1). aErrorbars indicate confidence limits.
HDRS scores significantly decreased after chrono­ Abbreviation: HDRS = Hamilton Depression Rating Scale.
therapeutics (Figure 1; main effect of time: F3,411 = 119.04,
P < .00001). from treatment and showed significantly decreased final HDRS
This effect was not influenced by previous history scores in respect to baseline whether they did (post hoc Newman-
of suicide attempts (2-way interaction of time and Keuls test, P = .024) or did not have (P = .001) a positive history
history: F3,411 = 0.55, P = .647; 3-way interaction of time, of suicide attempts.
response to treatment, and history: F3,411 = 1.42, P = .237). The significant HDRS decrease included the suicide item
Remarkably, however, nonresponders achieved a benefit (Figure 2; main effect of time: F3,411 = 42.78, P < .00001).

© 2013 COPYRIGHT PHYSICIANS POSTGRADUATE PRESS, INC. NOT FOR DISTRIBUTION, DISPLAY, OR COMMERCIAL PURPOSES.
J Clin Psychiatry e3
Benedetti et al

Figure 2. Pattern of Change of the HDRS Suicide Item


Responders with a positive suicide history showed
Scores During Treatment in Patients Divided According to higher baseline levels and a bigger decrease after the
Chronotherapeutic Treatment Response and to Previous History first total sleep deprivation + light therapy treatment
of Suicide Attempts (F3,291 = 6.31, P = .0004). Nonresponders, however,
who achieved a smaller, although significant benefit

Re lee p on

Re lee p on

p on

Re lee p on

Re lee p on

p on
from treatment (effect of time: F3,120 = 6.55, P = .0004),
l S Slee ati

l S Slee ati

ee ti

ee ti

ee ti

ee ti
Sl iva

ta y Sl riva

ta y Sl riva

Sl iva
iv

iv
To ery epr

To ery epr

ry epr

ry epr
r ep

r ep
showed a similar significant decrease in HDRS score
co p D

co p D

co p D

co p D

co p D

co p D
Re lee

Re lee
ve

ve

ve

ve
independent of history of previous suicide attempts
S

lS

lS
v

v
al

al
ta

ta
t

t
To

To

To

To
1.8
Positive history (n = 21) Positive history (n = 11)
(post hoc Newman-Keuls test: positive history, P = .039;
1.6 Negative history (n = 78) Negative history (n = 31) negative history, P = .004).
HDRS Suicide Item Score, Mean

1.4 A subgroup of 16 patients (11.4%) reported current


1.2 suicidal thinking or planning (HDRS suicide item score
1.0 ≥ 2) and showed the same trend of decrease in HDRS
0.8 suicide scores observed in the whole sample (effect
0.6 of time: F3,42 = 20.70, P < .000001) (Figure 3), with a
0.4 significant final benefit for nonresponders too (post hoc
0.2
Newman-Keuls test: P = .002 in respect to baseline). None
0.0
of the patients’ scores on the suicide item decreased.
Compared with patients who lacked current suicidal
0 1 2 3 4 5 6 0 1 2 3 4 5 6 planning, suicidal patients showed marginally worse
Time, d Time, d overall HDRS scores (F1,139 = 3.93, P = .049) due to
Responders (n = 99) Nonresponders (n = 42) significantly worse scores at baseline (day 0: β = .233,
aError
bars indicate confidence limits. t = 2.82, P = .005), but not during and after treatment
Abbreviation: HDRS = Hamilton Depression Rating Scale.
(day 1: t = 1.25, P = .21; day 2: t = 1.35, P = .18; day 6:
t = 1.30, P = .20).
Self-ratings of depression (13-item BDI) confirmed
Figure 3. Pattern of Change of Depression Severity (HDRS these same effects. Correlations between BDI and
scores) During Chronotherapeutic Treatment in Patients Divided the corresponding HDRS scores were all significant
According to the Presence or Absence of Current Suicidal
Thinking or Planninga (r = 0.23, 0.48, 0.49, and 0.61, at days 1, 2, 3, and 7,
respectively). A complete set of 4 daily ratings was
obtained in 109 of 141 patients. Patterns of change over
n

n
io

io

io
at

at

at
iv

iv

iv

time were closely similar to those observed with HDRS:


pr

pr

pr
ep

ep

ep
De

De

De
le

le

le
yS

yS

yS
p

a highly significant decrease in depression severity after


ee

ee

ee
r

r
ve

ve

ve
Sl

Sl

Sl
co

co

co
al

al

al

treatment (F3,324 = 36.36, P < .000001), with a significant


t

t
Re

Re

Re
To

To

To

26
24 Current suicidal thinking (n = 16) amelioration of symptoms also found in nonresponders
HDRS suicide item score < 2
22 (n = 125) (post hoc Newman-Keuls, P = .0007); no influence of
20 previous history of suicide attempts (interaction with
HDRS Score, Mean

18 time: F3,321 = 0.85, P = .469) nor of current suicidality


16 (F3,321 = 1.59, P = .192); and, most important, a highly
14 significant decrease of the suicide item (F3,342 = 9.99,
12 P = .000002), again involving a significant decrease in
10 nonresponders (post hoc Newman-Keuls, P = .0006).
8
6
Effect of Lithium
4
0 1 2 3 4 5 6 Ongoing lithium treatment conferred some
Time, d advantage in respect to newly started lithium (Figure
aErrorbars indicate confidence limits. 4), with a significant main effect on global HDRS scores
Abbreviation: HDRS = Hamilton Depression Rating Scale. (F1,139 = 4.18, P = .043) due to a better improvement after
the first total sleep deprivation + light therapy cycle (day
2: β = .169, t = 2.03, P = .045) but with similar final scores
Previous history of suicide attempts was associated with worse (day 7: β = .087, t = 1.03, P = .303).
suicide scores at baseline (β = .248, t = 2.84, P = .005) and did not History of drug resistance did not significantly
interact with time in influencing the decrease of HDRS suicide influence the pattern of decrease of HDRS scores, either
scores (F3,411 = 0.77, P = .512), but it significantly interacted when comparing drug-resistant and non–drug-resistant
with time and response (F3,411 = 3.42, P = .017). This differential patients (F3,417 = 1.46, P = .223) or when dividing the
result occurred because the interaction of time and history was drug-resistant patients into 3 classes (stage 1, stage 2,
significant among responders but not among nonresponders. stage 3 or higher: F9,411 = 1.31, P = .232).

© 2013 COPYRIGHT PHYSICIANS POSTGRADUATE PRESS, INC. NOT FOR DISTRIBUTION, DISPLAY, OR COMMERCIAL
e4
PURPOSES.
J Clin Psychiatry
Efficacy of Chronotherapy on Suicidality

Figure 4. Pattern of Change of Depression Severity (HDRS


Confirming the literature, our results showed that a
scores) During Chronotherapeutic Treatment in Patients positive history compared to a negative history of suicidal
Divided According to the Status of Lithium Medication acts was associated with more early stress32,52 and with
(ongoing or newly started)a worse current suicidality.53 Nevertheless, 66% of patients
with a positive history responded with a rapid drop in

n
io

io

io
suicidal symptoms. Remarkably, suicidality also decreased

at

at

at
iv

iv

iv
pr

pr

pr
p

p
De

De

De
ee

ee
in nonresponders, who then obtained a substantial benefit

e
Sl

Sl

Sl
ep

ep

ep
ry

ry

ry
le

le

le
ve

ve

ve
despite not achieving final response.
S

lS
co

co

co
al

al

ta
t

t
Re

Re

Re
To

To

To
22
Ongoing lithium treatment (n = 49)
These effects on suicidal thinking confirm previous studies
20
Newly started lithium (n = 92) showing that sleep deprivation rapidly reverts the baseline
18 mood-congruent cognitive biases toward negative stimuli
in depression.54,55 Cognitive distortions include pessimism
HDRS Score, Mean

16

14 and self-deprecatory and self-accusatory thoughts. We


12 showed that effective antidepressant chronotherapeutics
normalized the reactivity of corticolimbic circuitries to
10
emotional and moral stimuli, thus restoring an efficient
8
top-down control on response to negative stimuli by cortical
6 regions involved in the generation and control of depressed
4 mood.56
0 1 2 3 4 5 6
Mechanisms of these rapid effects involve multitarget
Time, d
actions on several neurobiological pathways.57 Total
aErrorbars indicate confidence limits. sleep deprivation acts on brain serotonin, dopamine,
Abbreviation: HDRS = Hamilton Depression Rating Scale.
norepinephrine, glutamate, and adenosine, and changes in
these signaling cascades are proportional to the observed
behavioral effects.58 These factors influence both behavior59
During the following month with lithium monotherapy, and brain metabolism and reactivity56,60,61 as well as a steady
21 of 99 responders showed signs of relapse and were increase of excitability of cortical circuits.62,63 Remarkably, a
administered an adjunctive antidepressant treatment by the comparable rapid effect on suicidality had been reported for
psychiatrists in charge, while the other 78 of 99 responders the glutamatergic N-methyl-d-aspartate (NMDA) receptor
continued to stay well and were discharged from the hospital antagonist ketamine.64–66 sleep deprivation itself alters the
with lithium therapy alone. circadian cyclicity of glutamatergic neurotransmission,63
with changes of glutamate/glutamine concentrations being
DISCUSSION proportional to mood improvement,67 and has been shown
The 1-week combination of lithium and chronotherapeutics to modify the expression of NMDA receptor subunits68
was followed by a 70% response rate in patients affected by and reduce NMDA sensitivity in model organisms.69
bipolar depression. Considering the 21% relapse rate in the Treatment response in this study was significant on day 1,
following weeks, this brings the success to 55.3% of patients but it continued to drop after treatment repetition, while,
achieving a sustained response. Consistent with previous in contrast, the suicidal ideation dropped to a very low level
studies,47 the rate of switch into mania was very low (1.4%) on day 1 and remained low thereafter (Table 1). This is
and manic symptoms rapidly disappeared. directly compatible with the effects reported after extremely
The rate of success is higher than that reported in previous low doses of ketamine.64–66 Changes of glutamatergic
studies of drug-resistant bipolar depressed patients (44% of neurotransmission could then be a core component of these
acute response, with only 40% of responders staying well after effects uniquely common to the 2 treatments.
1 month).29 The improvement in success rate from 17.6% Lithium could potentiate these mechanisms by acting
to 55.3% could be due to 2 methodological innovations: on signaling cascades downstream the monoaminergic
starting lithium together with chronotherapeutics in patients pathways,70,71 by promoting synaptic plasticity,72 and
not treated with lithium and prolonging light therapy for 1 by overcoming the detrimental effects of genetic factors
week after the acute response. Lithium promotes response in associated with treatment failure and suicide.32,36 Lithium
nonresponders48 and can overcome the detrimental effects of protects against both the reduction of gray matter volumes
biological factors hampering response to antidepressants.36,49 associated with suicidal behavior in bipolar disorder32 and
Light therapy promotes response in drug-resistant the disruption of white matter connecting corticolimbic
depression.50,51 Both lithium and light therapy are synergistic circuitries. 73,74 Bipolar disorder, 75 drug-resistant
with sleep deprivation,30 and could then interact to sustain its depression,76 and suicide77 have been associated with a
effects. The decrease in suicide scores was immediate, soon proinflammatory state and with an abnormal activation of
after the first total sleep deprivation + light therapy, thus brain microglia and of circulating monocytes,78 and again
suggesting a direct therapeutic effect of total sleep deprivation both sleep manipulations79 and lithium80 target inflammatory
(and not of recovery sleep) on this symptom (Figures 2–3). mechanisms associated with depression.81,82

© 2013 COPYRIGHT PHYSICIANS POSTGRADUATE PRESS, INC. NOT FOR DISTRIBUTION, DISPLAY, OR COMMERCIAL PURPOSES.
J Clin Psychiatry e5
Benedetti et al

The clinical relevance of the above is remarkable in light depressive suicidality and prompt antidepressant response
of the debated issue of how to treat suicidality in depression. in drug-resistant major depression in the course of bipolar
In other psychiatric conditions, such as schizophrenia, a disorder. This study adds new evidence to warrant a role
protective effect of adequate and effective antipsychotic for chronotherapeutics as a first-line treatment for bipolar
medication against suicide has been repeatedly confirmed.83 depression.
The analysis of proprietary data submitted to the US Food Drug names: ketamine (Ketalar and others), lithium (Lithobid and others).
and Drug Administration suggests instead that the net effect Author affiliations: Department of Clinical Neurosciences, Scientific
of antidepressant drugs, compared with placebo, could be Institute and University Vita-Salute San Raffaele, Milan, Italy.
Potential conflicts of interest: The authors have no financial conflicts of
negative among adults aged under 25 years, neutral on suicidal interest to disclose.
behavior but possibly protective for suicidal ideation in adults Funding/support: The study was partly funded by the European Union (FP7
aged 25–64 years, and protective against both suicidality and grant 222963-MOODINFLAME).
Role of the sponsor: The study sponsor had no role in the design and conduct
suicidal behavior in those aged more than 65 years.12 In the of the study; collection, management, analysis, and interpretation of the data;
present study, we observed no suicidal behavior (completed and preparation, review, or approval of the manuscript.
suicide, attempted suicide, or preparatory acts), and no
REFERENCES
worsening of suicidal ideation among participants, who
were aged a mean (SD) of 47.16 (11.50) years (range, 19–71).   1. Post RM, Leverich GS, Altshuler LL, et al. Differential clinical characteristics,
medication usage, and treatment response of bipolar disorder in the US
Only 3 participants were aged < 26 years, thus preventing versus The Netherlands and Germany. Int Clin Psychopharmacol.
conclusions on the effects of chronotherapeutics on suicide 2011;26(2):96–106. doi:10.97/YICb3e284019PuMd
in younger adults, but our observation of protective effects   2. Altshuler LL, Kupka RW, Hellemann G, et al. Gender and depressive
symptoms in 711 patients with bipolar disorder evaluated prospectively in the
in the overall sample confirms its central role as a rapid first- Stanley Foundation bipolar treatment outcome network. Am J Psychiatry.
line option in bipolar depression.27 Again, lithium could 2010;167(6):708–715. doi:10.76/apj29105PubMed
favorably interact with chronotherapeutics in limiting the   3. Kupka RW, Altshuler LL, Nolen WA, et al. Three times more days depressed
than manic or hypomanic in both bipolar I and bipolar II disorder. Bipolar
suicide symptoms because it has a clear protective effect on Disord. 2007;9(5):531–535. doi:10./j39-5682704.xPubMed
suicide behaviors in bipolar disorder15,16 and it can sustain   4. Post RM, Leverich GS, Altshuler LL, et al. Relationship of prior
and enhance the antidepressant effect of sleep deprivation. antidepressant exposure to long-term prospective outcome in bipolar I
disorder outpatients. J Clin Psychiatry. 2012;73(7):924–930. doi:10.48/JCPm7396ubMed
It has been proposed that lithium can act on suicide by   5. Chen YW, Dilsaver SC. Lifetime rates of suicide attempts among subjects
decreasing impulsivity, aggression, or decision-making with bipolar and unipolar disorders relative to subjects with other Axis I
deficits, which are endophenotypes intermediate between disorders. Biol Psychiatry. 1996;39(10):896–899. doi:10.6/-32(95) PubMed
  6. Leverich GS, Altshuler LL, Frye MA, et al. Factors associated with suicide
suicidal thoughts and behaviors84,85; our data suggest that, at attempts in 648 patients with bipolar disorder in the Stanley Foundation
least when sustaining effective chronotherapeutics, lithium Bipolar Network. J Clin Psychiatry. 2003;64(5):506–515. doi:10.48/JCPv6n53ubMed
could also help to decrease suicidal thoughts as rated on the   7. Osby U, Brandt L, Correia N, et al. Excess mortality in bipolar and unipolar
disorder in Sweden. Arch Gen Psychiatry. 2001;58(9):844–850. doi:10./archpsy5894PubMed
BDI and HDRS.   8. Jamison KR. Suicide and bipolar disorders. Ann N Y Acad Sci. 1986;487
A major caveat when using chronotherapeutics in clinical (1 Psychobiology):301–315. doi:10./j749-6328tb0.xPuMed
settings comes from the possibility of depressive relapse in   9. Partonen T, Sihvo S, Lönnqvist JK. Patients excluded from an antidepressant
efficacy trial. J Clin Psychiatry. 1996;57(12):572–575. doi:10.48/JCPv57n23ubMed
the first month after chronotherapeutics. This is all but 10. Zimmerman M, Mattia JI, Posternak MA. Are subjects in pharmacological
unexpected in patients with bipolar disorder, who present treatment trials of depression representative of patients in routine clinical
depression and depressive cycling as a substantial problem in practice? Am J Psychiatry. 2002;159(3):469–473. doi:10.76/apj5934PubMed
11. Tiihonen J, Lönnqvist J, Wahlbeck K, et al. Antidepressants and the risk of
some two-thirds of intensively treated bipolar outpatients86 suicide, attempted suicide, and overall mortality in a nationwide cohort. Arch
and who remain well on antidepressants for more than 2 Gen Psychiatry. 2006;63(12):1358–1367. doi:10./archpsy63258PubMed
months at a very low rate,86 with further discontinuation 12. Stone M, Laughren T, Jones ML, et al. Risk of suicidality in clinical trials of
antidepressants in adults: analysis of proprietary data submitted to US Food
being associated with a further substantially increased risk and Drug Administration. BMJ. 2009;339:b2880. doi:10.36/bmj28PuMed
of depressive relapse.22 A careful monitoring of mood in the 13. Walsh BT, Seidman SN, Sysko R, et al. Placebo response in studies of major
weeks after acute response to chronotherapeutics is then depression: variable, substantial, and growing. JAMA.
2002;287(14):1840–1847. doi:10./jam2874PubMed
advised24 in order to promptly start that complex pattern of 14. Turner EH, Matthews AM, Linardatos E, et al. Selective publication of
medication trials, which chronotherapeutics could not avoid antidepressant trials and its influence on apparent efficacy. N Engl J Med.
in these cases.23 2008;358(3):252–260. doi:10.56/NEJMsa79Pubed
15. Baldessarini RJ, Tondo L, Hennen J. Lithium treatment and suicide risk in
Limitations of the present study include lack of placebo major affective disorders: update and new findings. J Clin Psychiatry.
control for the chronotherapeutic procedure, lack of 2003;64(suppl 5):44–52.PubMed
assessment of lifetime medications and of their possible 16. Goodwin FK, Fireman B, Simon GE, et al. Suicide risk in bipolar disorder
during treatment with lithium and divalproex. JAMA.
effects on the current status, absence of drug-naive patients, 2003;290(11):1467–1473. doi:10./jam29467PubMed
absence of evaluation for lifetime compliance, absence 17. Keilp JG, Grunebaum MF, Gorlyn M, et al. Suicidal ideation and the
of testing of interrater reliability, and the low number of subjective aspects of depression. J Affect Disord. 2012;140(1):75–81. doi:10.6/ja2 45PubMed
18. Grunebaum MF, Keilp J, Li S, et al. Symptom components of standard
patients who reported current suicidal thinking or planning, depression scales and past suicidal behavior. J Affect Disord.
which prevented the study of interactions (eg, the possible 2005;87(1):73–82. doi:10.6/ja253PubMed
potentiating effect of lithium salts) in this subgroup. 19. Thase ME. Antidepressant combinations: widely used, but far from
empirically validated. Can J Psychiatry. 2011;56(6):317–323.PubMed
In conclusion, the combination of total sleep deprivation, 20. Frye MA, Helleman G, McElroy SL, et al. Correlates of treatment-emergent
light therapy, and lithium was able to rapidly decrease mania associated with antidepressant treatment in bipolar depression. Am J

© 2013 COPYRIGHT PHYSICIANS POSTGRADUATE PRESS, INC. NOT FOR DISTRIBUTION, DISPLAY, OR COMMERCIAL
e6
PURPOSES.
J Clin Psychiatry
Efficacy of Chronotherapy on Suicidality

Psychiatry. 2009;166(2):164–172. doi:10.76/apj2830PubMed 49. Benedetti F, Colombo C, Serretti A, et al. Antidepressant effects of light
21. Gao K, Kemp DE, Ganocy SJ, et al. Treatment-emergent mania/hypomania therapy combined with sleep deprivation are influenced by a functional
during antidepressant monotherapy in patients with rapid cycling bipolar polymorphism within the promoter of the serotonin transporter gene. Biol
disorder. Bipolar Disord. 2008;10(8):907–915. doi:10./j39-568207.xPubMed Psychiatry. 2003;54(7):687–692. doi:10.6/S-32()1894PubMed
22. Altshuler L, Suppes T, Black D, et al. Impact of antidepressant discontinuation 50. Prasko J, Brunovsky M, Latalova K, et al. Augmentation of antidepressants
after acute bipolar depression remission on rates of depressive relapse at with bright light therapy in patients with comorbid depression and borderline
1-year follow-up. Am J Psychiatry. 2003;160(7):1252–1262. doi:10.76/apj 25PubMed personality disorder. Biomed Pap Med Fac Univ Palacky Olomouc Czech
23. Post RM, Altshuler LL, Frye MA, et al. Complexity of pharmacologic Repub. 2010;154(4):355–361. doi:10.57/bp23PuMed
treatment required for sustained improvement in outpatients with bipolar 51. Goel N, Terman M, Terman JS, et al. Controlled trial of bright light and
disorder. J Clin Psychiatry. 2010;71(9):1176–1186, quiz 1252–1253. doi:10.48/JCPmyelubMd negative air ions for chronic depression. Psychol Med. 2005;35(7):945–955. doi:10.7/S3295 PubMed
24. Benedetti F, Colombo C. Sleep deprivation in mood disorders. 52. Labonte B, Turecki G. Epigenetic Effects of Childhood Adversity in the Brain
Neuropsychobiology. 2011;64(3):141–151. doi:10.59/32847PubMed and Suicide Risk. In: Dwivedi Y, ed. The Neurobiological Basis of Suicide. Boca
25. Benedetti F, Barbini B, Colombo C, et al. Chronotherapeutics in a psychiatric Raton, FL: CRC Press; 2012. doi:10.2/b5-4
ward. Sleep Med Rev. 2007;11(6):509–522. doi:10.6/jsmrv274PubMed 53. Oquendo MA, Galfalvy H, Russo S, et al. Prospective study of clinical
26. Wirz-Justice A, Benedetti F, Berger M, et al. Chronotherapeutics (light and predictors of suicidal acts after a major depressive episode in patients with
wake therapy) in affective disorders. Psychol Med. 2005;35(7):939–944. doi:10.7/S32954XPubMed major depressive disorder or bipolar disorder. Am J Psychiatry.
27. Wirz-Justice A, Benedetti F, Terman M. Chronotherapeutics for Affective 2004;161(8):1433–1441. doi:10.76/apj843PubMed
Disorders. A Clinician’s Manual for Light and Wake Therapy. Basel, 54. Baving L, Maes H, Bohus M, et al. Can negative self-schemes in depressives
Switzerland: Karger; 2009. be altered through sleep deprivation? J Affect Disord. 1997;42(2–3):93–101. doi:10.6/S5-327(9)140PubMed
28. Benedetti F, Dallaspezia S, Fulgosi MC, et al. Phase advance is an actimetric 55. Benedetti F, Barbini B, Florita M, et al. Rapid improvement in information
correlate of antidepressant response to sleep deprivation and light therapy in processing after sleep deprivation and sleep phase-advance in bipolar
bipolar depression. Chronobiol Int. 2007;24(5):921–937. doi:10.8/742569PubMed depression. Clinical Neuropsychiatry. 2005;2(3):180–182.
29. Benedetti F, Barbini B, Fulgosi MC, et al. Combined total sleep deprivation 56. Benedetti F, Bernasconi A, Blasi V, et al. Neural and genetic correlates of
and light therapy in the treatment of drug-resistant bipolar depression: acute antidepressant response to sleep deprivation: a functional magnetic
response and long-term remission rates. J Clin Psychiatry. resonance imaging study of moral valence decision in bipolar depression.
2005;66(12):1535–1540. doi:10.48/JCPv6n27ubMed Arch Gen Psychiatry. 2007;64(2):179–187. doi:10./archpsy64279PubMed
30. Colombo C, Lucca A, Benedetti F, et al. Total sleep deprivation combined 57. Millan MJ. Multi-target strategies for the improved treatment of depressive
with lithium and light therapy in the treatment of bipolar depression: states: conceptual foundations and neuronal substrates, drug discovery and
replication of main effects and interaction. Psychiatry Res. 2000;95(1):43–53. doi:10.6/S5-78()14PubMed therapeutic application. Pharmacol Ther. 2006;110(2):135–370. doi:10.6/jpharmte251.06PubMd
31. First MB, Spitzer RL, Gibbon M, et al. Structured Clinical Interview for DSM- 58. Benedetti F, Smeraldi E. Neuroimaging and genetics of antidepressant
IV Axis I Disorders, Research Version, Patient Edition. New York, NY: response to sleep deprivation: implications for drug development. Curr
Biometric Research, New York State Psychiatric Institute; 1997. Pharm Des. 2009;15(22):2637–2649. doi:10.274/38695PubMed
32. Benedetti F, Radaelli D, Poletti S, et al. Opposite effects of suicidality and 59. Benedetti F, Dallaspezia S, Lorenzi C, et al. Gene-gene interaction of glycogen
lithium on gray matter volumes in bipolar depression. J Affect Disord. synthase kinase 3-β and serotonin transporter on human antidepressant
2011;135(1–3):139–147. doi:10.6/ja27PubMed response to sleep deprivation. J Affect Disord. 2012;136(3):514–519. doi:10.6/ja2 39PubMed
33. Hamilton M. Development of a rating scale for primary depressive illness. Br 60. Ebert D, Berger M. Neurobiological similarities in antidepressant sleep
J Soc Clin Psychol. 1967;6(4):278–296. doi:10./j24-8697tb053.xPuMed deprivation and psychostimulant use: a psychostimulant theory of
34. Benedetti F, Colombo C, Barbini B, et al. Ongoing lithium treatment prevents antidepressant sleep deprivation. Psychopharmacology (Berl).
relapse after total sleep deprivation. J Clin Psychopharmacol. 1998;140(1):1–10. doi:10.7/s235PubMed
1999;19(3):240–245. doi:10.97/4-60 7PubMed 61. Wu JC, Buchsbaum M, Bunney WE Jr. Clinical neurochemical implications
35. Baxter LR Jr. Can lithium carbonate prolong the antidepressant effect of sleep of sleep deprivation’s effects on the anterior cingulate of depressed
deprivation? Arch Gen Psychiatry. 1985;42(6):635. doi:10./archpsy9857021PubMed responders. Neuropsychopharmacology. 2001;25(suppl):S74–S78. doi:10.6/S893-X()0PubMed
36. Benedetti F, Barbini B, Bernasconi A, et al. Lithium overcomes the influence 62. Huber R, Mäki H, Rosanova M, et al. Human cortical excitability increases
of 5-HTTLPR gene polymorphism on antidepressant response to sleep with time awake. Cereb Cortex. 2013;23(2):332–338.PubMed
deprivation. J Clin Psychopharmacol. 2008;28(2):249–251. doi:10.97/JCPb3e8641uMd 63. Vyazovskiy VV, Cirelli C, Pfister-Genskow M, et al. Molecular and
37. Leibenluft E, Moul DE, Schwartz PJ, et al. A clinical trial of sleep deprivation electrophysiological evidence for net synaptic potentiation in wake and
in combination with antidepressant medication. Psychiatry Res. depression in sleep. Nat Neurosci. 2008;11(2):200–208. doi:10.38/n25PubMed
1993;46(3):213–227. doi:10.6/5-78(93)04PubMed 64. Larkin GL, Beautrais AL. A preliminary naturalistic study of low-dose
38. Beck AT, Beck RW. Screening depressed patients in family practice: a rapid ketamine for depression and suicide ideation in the emergency department.
technic. Postgrad Med. 1972;52(6):81–85.PubMed Int J Neuropsychopharmacol. 2011;14(8):1127–1131. doi:10.7/S465 29PubMed
39. Desseilles M, Perroud N, Guillaume S, et al. Is it valid to measure suicidal 65. DiazGranados N, Ibrahim LA, Brutsche NE, et al. Rapid resolution of suicidal
ideation by depression rating scales? J Affect Disord. 2012;136(3):398–404. doi:10.6/ja2 3PubMed ideation after a single infusion of an N-methyl-d-aspartate antagonist in
40. Valtonen HM, Suominen K, Sokero P, et al. How suicidal bipolar patients are patients with treatment-resistant major depressive disorder. J Clin Psychiatry.
depends on how suicidal ideation is defined. J Affect Disord. 2010;71(12):1605–1611. doi:10.48/JCP9m5327bluMed
2009;118(1–3):48–54. doi:10.6/ja298PubMed 66. Price RB, Nock MK, Charney DS, et al. Effects of intravenous ketamine on
41. Thase ME, Rush AJ. When at first you don’t succeed: sequential strategies for explicit and implicit measures of suicidality in treatment-resistant depression.
antidepressant nonresponders. J Clin Psychiatry. 1997;58(suppl 13):23–29.PubMed Biol Psychiatry. 2009;66(5):522–526. doi:10.6/jbpsych2940PuMed
42. Holmes TH, Rahe RH. The Social Readjustment Rating Scale. J Psychosom 67. Benedetti F, Calabrese G, Bernasconi A, et al. Spectroscopic correlates of
Res. 1967;11(2):213–218. doi:10.6/2-39(7)104PubMed antidepressant response to sleep deprivation and light therapy: a 3.0 Tesla
43. Dohrenwend BP. Inventorying stressful life events as risk factors for study of bipolar depression. Psychiatry Res. 2009;173(3):238–242. doi:10.6/jpscyhren280.4PubMd
psychopathology: toward resolution of the problem of intracategory 68. Park HJ, Kang WS, Paik JW, et al. Effect of valproic acid through regulation of
variability. Psychol Bull. 2006;132(3):477–495. doi:10.37/-29 4PubMed NMDA receptor-ERK signaling in sleep deprivation rats. J Mol Neurosci.
44. McCulloch CE, Searle SR, Neuhaus JM. Generalized, Linear, and Mixed 2012;47(3):554–558. doi:10.7/s23-965PubMed
Models. 2nd ed. New York, NY: John Wiley & Sons; 2008. 69. Novati A, Hulshof HJ, Granic I, et al. Chronic partial sleep deprivation
45. Timm N, Kim K. Univariate and Multivariate General Linear Models: Theory reduces brain sensitivity to glutamate N-methyl-D-aspartate receptor-
and Applications With SAS. 2nd ed. Berlin, Heidelberg, Germany: Springer; mediated neurotoxicity. J Sleep Res. 2012;21(1):3–9. doi:10./j365-2890.xPubMed
2006. 70. Beaulieu JM, Caron MG. Looking at lithium: molecular moods and complex
46. Hill T, Lewicki P. Statistics: Methods And Applications. A Comprehensive behaviour. Mol Interv. 2008;8(5):230–241. doi:10.24/m85PubMed
Reference For Science, Industry, And Data Mining, Chapter 18. Tulsa, OK: 71. Beaulieu JM, Gainetdinov RR, Caron MG. Akt/GSK3 signaling in the action
General Linear Models, StatSoft; 2006:245–276. of psychotropic drugs. Annu Rev Pharmacol Toxicol. 2009;49(1):327–347. doi:10.46/anurevphmtox018.4563PubMed
47. Colombo C, Benedetti F, Barbini B, et al. Rate of switch from depression into 72. Schloesser RJ, Martinowich K, Manji HK. Mood-stabilizing drugs:
mania after therapeutic sleep deprivation in bipolar depression. Psychiatry mechanisms of action. Trends Neurosci. 2012;35(1):36–46. doi:10.6/jtns2 9PubMed
Res. 1999;86(3):267–270. doi:10.6/S5-78(9)03PubMed 73. Benedetti F, Absinta M, Rocca MA, et al. Tract-specific white matter
48. Bschor T, Bauer M. Efficacy and mechanisms of action of lithium structural disruption in patients with bipolar disorder. Bipolar Disord.
augmentation in refractory major depression. Curr Pharm Des. 2011;13(4):414–424. doi:10./j39-56820.xPubMed
2006;12(23):2985–2992. doi:10.274/386 950PubMed 74. Benedetti F, Bollettini I, Barberi I, et al. Lithium and GSK3-β promoter gene

© 2013 COPYRIGHT PHYSICIANS POSTGRADUATE PRESS, INC. NOT FOR DISTRIBUTION, DISPLAY, OR COMMERCIAL PURPOSES.
J Clin Psychiatry e7
Benedetti et al

variants influence white matter microstructure in bipolar disorder. 81. Voderholzer U, Fiebich BL, Dersch R, et al. Effects of sleep deprivation on
Neuropsychopharmacology. 2013;38(2):313–327. doi:10.38/np27PubMed nocturnal cytokine concentrations in depressed patients and healthy control
75. Padmos RC, Hillegers MH, Knijff EM, et al. A discriminating messenger RNA subjects. J Neuropsychiatry Clin Neurosci. 2012;24(3):354–366. doi:10.76/apneursych1042PbMed
signature for bipolar disorder formed by an aberrant expression of 82. Voderholzer U, Hohagen F, Klein T, et al. Impact of sleep deprivation and
inflammatory genes in monocytes. Arch Gen Psychiatry. 2008;65(4):395–407. doi:10./archpsy65439PubMed subsequent recovery sleep on cortisol in unmedicated depressed patients. Am
76. Lanquillon S, Krieg JC, Bening-Abu-Shach U, et al. Cytokine production and J Psychiatry. 2004;161(8):1404–1410. doi:10.76/apj84PubMed
treatment response in major depressive disorder. Neuropsychopharmacology. 83. Heilä H, Isometsä ET, Henriksson MM, et al. Suicide victims with
2000;22(4):370–379. doi:10.6/S893-X()147PubMed schizophrenia in different treatment phases and adequacy of antipsychotic
77. Steiner J, Bielau H, Brisch R, et al. Immunological aspects in the neurobiology medication. J Clin Psychiatry. 1999;60(3):200–208. doi:10.48/JCPv6n3ubMed
of suicide: elevated microglial density in schizophrenia and depression is 84. Gould TD, Can A, Gottesman II, et al. Differential lithium efficacy in
associated with suicide. J Psychiatr Res. 2008;42(2):151–157. doi:10.6/jpsychre210.3PubMd reducing suicidal behaviors compared with suicidal thoughts. Am J
78. Drexhage RC, Knijff EM, Padmos RC, et al. The mononuclear phagocyte Psychiatry. 2012;169(1):98–99, author reply 99. doi:10.76/apj2 810PubMed
system and its cytokine inflammatory networks in schizophrenia and bipolar 85. Kovacsics CE, Gottesman II, Gould TD. Lithium’s antisuicidal efficacy:
disorder. Expert Rev Neurother. 2010;10(1):59–76. doi:10.586/ern94PubMd elucidation of neurobiological targets using endophenotype strategies. Annu
79. Imeri L, Opp MR. How (and why) the immune system makes us sleep. Nat Rev Pharmacol Toxicol. 2009;49(1):175–198. doi:10.46/anurevphmtox018.457PubMed
Rev Neurosci. 2009;10(3):199–210. doi:10.38/nr2576PubMed 86. Post RM, Leverich GS, Nolen WA, et al; Stanley Foundation Bipolar Network.
80. Jope RS, Yuskaitis CJ, Beurel E. Glycogen synthase kinase-3 (GSK3): A re-evaluation of the role of antidepressants in the treatment of bipolar
inflammation, diseases, and therapeutics. Neurochem Res. depression: data from the Stanley Foundation Bipolar Network. Bipolar
2007;32(4–5):577–595. doi:10.7/s64-9285PubMed Disord. 2003;5(6):396–406. doi:10.46/j39-5820.xPubMed

© 2013 COPYRIGHT PHYSICIANS POSTGRADUATE PRESS, INC. NOT FOR DISTRIBUTION, DISPLAY, OR COMMERCIAL
e8
PURPOSES.
J Clin Psychiatry